Track Lanes Discussion

[vitalker]

Originally Posted by norbury brook

I'm late to this thread so forgive me my qusetion:

Is there a Reason you just don't copy Cubase/Logic/Studio 1 takes/comping systems?

They work really well and aren't complicated.


is there some kind of backwards compatibility issue stopping this?

By Reason you mean DAW? Are you asking us, the users?
Wouldn't it be a law problem, if they copy features? Also how do you expect the devs to copy anything? Firstly they'd need to get a copy and then make reverse-engineering of it. Otherwise it wouldn't be a copy, would it?
The devs just don't use other DAWs and make features how they like and maybe even better than other DAWs. It's like asking an artist to draw same picture.

[sockmonkey72]

Originally Posted by norbury brook

Is there a Reason you just don't copy Cubase/Logic/Studio 1 takes/comping systems?

I'm going to take a wild guess and go with "because REAPER isn't some copy/paste best-of DAW and has its own internal logic, history, inertia and, to some extent ego investment on the part of the team".

Also, White Tie reviewed all of those DAWs and didn't think they got it right.

From a technology point of view, there's absolutely no reason to believe that you can simply (and successfully) transplant designs into unrelated software, without a major upheaval in the code base. To the point where it might be more cost-effective to start over from scratch.

[ferropop]

It's less about copying designs and more about absorbing workflows in other DAWs - attempting to achieve an optimal end result in the most fluid way possible. The many flavours of comping/lanes that exist across DAWs all offer a different journey of failures/successes to study and draw influence from. To ignore them would be a mindboggling misstep.

[vitalker]

Originally Posted by ferropop

It's less about copying designs and more about absorbing workflows in other DAWs - attempting to achieve an optimal end result in the most fluid way possible. The many flavours of comping/lanes that exist across DAWs all offer a different journey of failures/successes to study and draw influence from. To ignore them would be a mindboggling misstep.

Can't say better!

[enroe]

Originally Posted by vitalker

It's like asking an artist to draw same picture.

I think that sums it up well.

[norbury brook]

Originally Posted by ferropop

It's less about copying designs and more about absorbing workflows in other DAWs - attempting to achieve an optimal end result in the most fluid way possible. The many flavours of comping/lanes that exist across DAWs all offer a different journey of failures/successes to study and draw influence from. To ignore them would be a mindboggling misstep.

exactly . I didn't mean literally copy their code!!!! I meant how they work. Logic,Cubase and Studio 1 are very similar in how they do comps/takes hence me using them as examples. The fact they are ll so similar would make one believe it's because it's proved to be a good way of doing things that the user base likes.

I think the drawing the same picture analogy isn't a very good one. I think a better one would be an artist seeing another artist use a new medium in a different way also starts to use that medium too but in his own way.

The takes/comping system is a workflow thing so you're copying a workflow.

If the answer is it's too difficult because of 'a,b, or c' that's fine I was just asking a question.



M

[Soli Deo Gloria]

Hi,


Excuse me if I'm rather lost here, regarding the track lane discussion... I'm looking for this feature in the latest pre (6.71 rc1), but I can't find it anywhere. I remember there were options when right-clicking on a track... Was this feature temporarily disabled or removed? Again, sorry if I haven't read any obvious news regarding this.

[vitalker]

Originally Posted by Soli Deo Gloria

Hi,


Excuse me if I'm rather lost here, regarding the track lane discussion... I'm looking for this feature in the latest pre (6.71 rc1), but I can't find it anywhere. I remember there were options when right-clicking on a track... Was this feature temporarily disabled or removed? Again, sorry if I haven't read any obvious news regarding this.

RC is release candidate, not pre-release. You should get last dev build.

[Soli Deo Gloria]

Oh, silly me!


Thanks so much...

[BartR]

Originally Posted by ferropop

It's less about copying designs and more about absorbing workflows in other DAWs - attempting to achieve an optimal end result in the most fluid way possible. The many flavours of comping/lanes that exist across DAWs all offer a different journey of failures/successes to study and draw influence from. To ignore them would be a mindboggling misstep.

Golden words

[bolgwrad]

The current behaviour for recording into a fixed lane track either adds a new lane at the bottom of the track,
or if you check 'record into lane' (my preference, the top lane),
all the other lanes are muted while recording. I don't think there's a native way to change this.

So, rather than having to put each previous take in a child track or whatever,
eg for BVs or alternate takes, when you want to 'record into lane' but hear the other lanes at the same time, here's a quick and dirty way.

Add an empty top lane (Alt+] here).
Select 'record into lane' on its header button, which persists.
Set the record mode to 'Time selection auto-punch' and draw a short time selection after the project region, where you won't be recording.

When you hit record, none of the other lanes will be muted until you reach the time selection (which you won't), but your take is recorded in the selected lane.
You don't see the peaks in real time, but it's a tidier way of dropping in small takes.

[vsthem]

Has there been any word on this? It seems to have been temporarily shelved, at least on the Pre-Releases. Maybe now that Razor Edit Groups are released (which are great btw), they'll get back to this?

[MonkeyBars]

Originally Posted by vsthem

Has there been any word on this? It seems to have been temporarily shelved, at least on the Pre-Releases.

Nope, fixed lanes have been in pre-releases constantly.

Play Markers dev is currently on hold, though. They announced that they would pick up work on them after deploying some prerequisites/dependencies/etc.

[vsthem]

I guess I meant track lanes as they relate to comping, which I don't think you can really do without play markers. I hope they get the dependencies worked on sooner than later. I wonder if this will also lead to native track versions. Thanks for the update!

[antiClick]

Originally Posted by ovnis

It exists some actions to move items inside item lanes, but we can't drag&drop items with the mouse (which is far easier than activating an action with a shortcut) inside a track with media lines auto! It would be nice to be able to use drag&drop.

what's that theme? Looks cool

[IrishRover79]

Originally Posted by vsthem

I guess I meant track lanes as they relate to comping, which I don't think you can really do without play markers. I hope they get the dependencies worked on sooner than later. I wonder if this will also lead to native track versions. Thanks for the update!

This comment has reminded me of a question I would be interested in people's views on. Do we want fixed lanes to serve both as an alternative to the take system which allows for swipe comping and as a way of having different versions of the same track, or do we want these things to be separate?

Personally, I see these as separate things. And this is despite my being a refugee from Pro Tools. In PT, the two things are blurred together in the playlists feature. But I actually think Steinberg and Apple have got things right in Cubase and Logic, where there is a distinction between take lanes (which enable swipe comping) and track alternatives / track versions.

As an example of how these things are different and can actually complement each other, imagine you're recording multiple takes of a lead vocal on different days and maybe even with different microphones. Fixed lanes would allow to organise your takes from each session (and hopefully enable swipe comping). Track versions would allow you to keep separate groups of takes recorded on different days or with different gear.

And the use of track versions extends well beyond this, as is shown by some excellent scripts already available via ReaPack. I particularly like and regularly use BirdBird's Track Versions.

So my ideal version of Reaper would have both native swipe comping via something like fixed item lanes and a true track versioning system, but they would be separate features. I'm curious what others think about this.

[MonkeyBars]

Originally Posted by IrishRover79

So my ideal version of Reaper would have both native swipe comping via something like fixed item lanes and a true track versioning system, but they would be separate features. I'm curious what others think about this.

I only just made the connection between fixed lanes and the possibilities they open up in regards to track versioning.

Regardless, it's clear from dev releases that lanes and takes are already designed as totally different logical levels that can interoperate. I.e., the items in fixed lanes can have takes just like any other items. That's pretty powerful!

[Gass n Klang]

Originally Posted by IrishRover79

So my ideal version of Reaper would have both native swipe comping via something like fixed item lanes and a true track versioning system, but they would be separate features. I'm curious what others think about this.

That would be ok, even if I find PTs system the best out there by far. Reapers lanes aren't very helpful at the moment:
1) you need to zoom vertically as soon as you record some takes to see anything
2) if you get 50 takes or so, you don't see anything no matter how your zoom level is
=> in my opinion take lanes don't work as such. They need to be some kind of subtracks that can be fully collapsed, reordered etc.
3) it's super difficult to quickly jump between takes
4) if you do let's say a full band recording in lanes, then some corrections in single instruments and after that another full take, how will takes not be messed up the way it currently works?

[bolgwrad]

Originally Posted by Gass n Klang

=> in my opinion take lanes don't work as such. They need to be some kind of subtracks that can be fully collapsed, reordered etc.

Of course you know this: If you record in a regular 'non-lane' track, then 'Take/explode takes into new tracks' you duplicate all the takes in a collapsible subfolder, which is more like the functionality I think you want, rather than what lanes do or don't do.

The current fixed-lane functionality (for comping) applicable to a child folder set would be nice.

[User41]

Originally Posted by IrishRover79

But I actually think Steinberg and Apple have got things right in Cubase and Logic, where there is a distinction between take lanes (which enable swipe comping) and track alternatives / track versions.

As an example of how these things are different and can actually complement each other, imagine you're recording multiple takes of a lead vocal on different days and maybe even with different microphones. Fixed lanes would allow to organise your takes from each session (and hopefully enable swipe comping). Track versions would allow you to keep separate groups of takes recorded on different days or with different gear.

And the use of track versions extends well beyond this, as is shown by some excellent scripts already available via ReaPack. I particularly like and regularly use BirdBird's Track Versions.

So my ideal version of Reaper would have both native swipe comping via something like fixed item lanes and a true track versioning system, but they would be separate features. I'm curious what others think about this.

I agree 100%. I don’t understand the nuts and bolts of how Reaper is currently coded (so take my opinion with a grain of salt), but if Reaper can’t easily figure out a way to best Logic and Cubase system, I’d prefer they mimic it. There is literally nothing I would change about the way takes and versions are handled in those DAWs.

[User41]

Originally Posted by bolgwrad

Of course you know this: If you record in a regular 'non-lane' track, then 'Take/explode takes into new tracks' you duplicate all the takes in a collapsible subfolder, which is more like the functionality I think you want, rather than what lanes do or don't do.

The current fixed-lane functionality (for comping) applicable to a child folder set would be nice.

That is pretty much my comping workflow. I record, then explode takes to new tracks. And then make those new tracks sub tracks to a main track (that has all the track FX, automation, etc). Then I use splits and mutes to assemble my comps.

It works for me but has a bunch of extra steps vs what I would do in PT, Logic or Cubase.

Oh yeah, you can’t uses “versioning” with my method without duping the entire track stack. Things get messy quickly here so if I am working on a project that needs extensive versioning, I sadly choose to leave Reaper and work in PT.

[vsthem]

Yeah, I'd love Track Versions as a separate entity from Track Lanes. I use Heda's Track Versions inside Track Inspector, but I wish Reaper had that function natively.

[Gass n Klang]

Originally Posted by bolgwrad

Of course you know this: If you record in a regular 'non-lane' track, then 'Take/explode takes into new tracks' you duplicate all the takes in a collapsible subfolder, which is more like the functionality I think you want, rather than what lanes do or don't do.

That‘s not really an alternative. As that may somehow work for single tracks, it doesn‘t for multitrack recordings/edits. The current lanes don‘t seem to have an advantage over other DAW‘s workflows out there, so their development should be canceled and re-done in my opinion. It doesn‘t cover what many people need (and what a major reason is, why PT for example is so popular).

[vitalker]

Originally Posted by Gass n Klang

That‘s not really an alternative. As that may somehow work for single tracks, it doesn‘t for multitrack recordings/edits. The current lanes don‘t seem to have an advantage over other DAW‘s workflows out there, so their development should be canceled and re-done in my opinion. It doesn‘t cover what many people need (and what a major reason is, why PT for example is so popular).

If you don't know how Reaper development works, then I'll tell you. If you see it goes in wrong direction, you can post your point of view in dev build threads, so everyone will see it. Track lanes are not ready yet, so released feature may look different at all, but sometimes feedback is needed.

[ferropop]

Reaper's current comping system is *very* flawed in very specific ways - but amazing in really specific other ways.

I hope they're able to retain/duplicate specifically the GREAT things about Reaper's current comping system, and fix the things that are just factually better in other modern workflows. (ie, swiping, no splits)


Current Breakdown as I see it:


1. Super quick cycling through active takes.

Good -- So Fast! Bam bam bam, cycle through with single keypresses.

Bad -- The boundaries are defined by SPLITS which is a mess, and the limitations of being in an Item Container.

Hope - Swipe Comping (using solo markers), which in essence is "setting the boundaries of the active take with a single mouse drag". Gorgeous. If we retained the fast cycling actions, along with fast boundary creation (swipe/solo-markers) we're golden!


2. Time-Selection Auto Punch

Good -- Want to punch in a phrase? Select time, record, BAM instant promotion of that new take to Active, with perfectly set boundaries and crossfades.

Bad -- Accessing the stuff "outside" of the split boundaries is damn near impossible. You have to move crossfades or item edges Just To See It, which means destroying all the crossfades/editing/Melodyne to the left or right of your punch -- Just to get to it! Also if you're punching ACROSS SPLITS your punch will have a trillion splits. BAD!!

Hope -- Lanes let you see the entire punch in (including preroll) -- the behaviour could be to simply "Automatically Set Solo Markers To Time Selection on Punch In" which would give us the exact same nice behaviour as currently, but The Entire Take is Visible And Available in its own discrete lane! And the punch IS WHOLE -- no splits -- because the Time Selection sets the new Solo Markers! Wanna include some words from the preroll? Just swipe it in - no splits, nothing hidden behind item boundaries, Done.


3. Collapsing to Audible Take

Good -- The comp is collapsible to hide the takes and only present what you hear

Bad -- It's an ITEM LEVEL collapse - which means Melodyne / VocAlign are being fed ALL THE TAKES, ALL THE SPLITS which makes the DAW bog down Immensely, and actually prevents proper use of VocAlign because it expects Contiguous items

Hope -- some organizational tools in the lane view, where you can have RAW TAKES, COMPS (some options), TUNED COMPS, VOCALIGNED COMPS, and then a specific lane that is "THE AUDIBLE LANE Exposed to the Track". And you can decide which lane that is, and that is the lane that collapses and is sent to ARA/the channel strip/our ears. And there be some basic tools like coloring, folders (organizational, not routing!), solo/mute.

ie, imagine making a "TAKES" folder consisting of a bunch of lanes with takes (this is where new recording passes go). Now Solo Markers on lanes inside this folder will swipe-promote TO THE TAKES FOLDER PARENT. Then you can promote that to the COMPS folder as a comp option. Want another comp? Just re-comp inside the TAKES folder (or duplicate the folder, and do a new comp), then promote the new comp to the COMPS folder As A New Lane. Now you have COMP OPTIONS. Then promote the COMP you want to the FINAL folder, which is what is audible and gets exposed to the outside world.


If we have enough tools to manage our raw takes vs the audible comp (coloring, folders, etc) - this almost negates the need for Versioning. Just promote different stuff from your COMPS folder, those are your versions.

[IrishRover79]

Originally Posted by ferropop

If we have enough tools to manage our raw takes vs the audible comp (coloring, folders, etc) - this almost negates the need for Versioning. Just promote different stuff from your COMPS folder, those are your versions.

I agree with you on what we want in terms of comping. But I'm not sure I agree that what you describe almost negates the need for versioning.

For me, a track version is a much broader concept with a lot more uses than lanes for different recordings of the same thing, which is all that fixed item lanes seem current to be designed for. It's also something which can be used throughout the production process, e.g. in composition, editing, mixing and mastering, not just in recording and comping.

Pro Tools smooshes the two things together. Its playlist feature seem originally to have been designed primarily with comping in mind but, as with everything, engineers gradually started to use it for other things, e.g.
-saving backups of raw material that they're editing in some way so they can always 'get back' to what the client sent
-cycling through different mix references or mix prints (to check the mix is heading in the right direction)
-storing different options for a part so that they can quickly audition those different options against the rest of the parts, e.g. different bass lines against the drums and guitars.

Steinberg and Apple then came along and saw that there were two separate things needed here. One was a way for users to do swipe comping of takes. The other was a way for users to be able to have different sets of material associated with a track and to be able to switch between them. Hence both Cubase and Logic have not only swipe comping on take lanes but Cubase has Track Versions, and Logic has Track Alternatives. I think the same pattern has also played out in Studio One.

I completely agree with you that the current take system is good for certain things and bad for others. I also know that just pointing to other DAWs and saying 'give us that!' is not likely to persuade the Reaper devs. I'm just raising the issue of lanes vs track versions because the devs have clearly listened to the clamour for swipe comping but if they are going to fiddle with Reaper's track system, I would like them to consider not just the need for swipe comping of takes but also the broader needs for true track versioning. It would be a real shame to find we're stuck with fixed item lanes as our 'solution' to track versioning, when that's not really what it's meant for.

The best thing would be for Reaper to find a way forward that transcends what's on offer in other DAWs. But if that's not possible, then I really think what's happened with Cubase, Logic and Studio One should not be ignored. The fact they all separate lanes from versions points to a real distinction they've all recognised in what modern DAW users need.

[MonkeyBars]

Originally Posted by IrishRover79

For me, a track version is a much broader concept with a lot more uses than lanes for different recordings of the same thing, which is all that fixed item lanes seem current to be designed for.

Not sure where you get that idea at all. Have you tried them out in the dev versions? Fixed lanes are totally agnostic, like subtracks. Nothing at all says "different recordings of the same thing" in their design to me.

[ferropop]

Agreed with your versioning comment, was trying to conceive of a brand new thing that leaves as much of Reaper alone and just adds to it, with as few ripples as possible.

Something like this:



...some kind of system where Swipe Comping promotes Up To The Parent, which could itself have a parent that promotes itself Up To Its Parent... some kind of Chain.

I'm making this up in realtime so forgive any cloudy thinking, but there's something to this.

ie, Swipe Comp from Raw Takes TO COMP 1 Lane. Or COMP2 Lane, etc

Then Swipe the comp you want from from COMPS up to MELODYNED which is where you'd do your tuning.

Swipe from MELODYNED up to the LEAD VOCAL Parent, which is AUDIBLE.

[ferropop]

^^^i'm just using tracks here to put together the concept -- again I imagine Lane Folders to be an Organizational thing, not a routing thing like normal folders -- with the idea being that The Top Folder is what is audible, and while comping there needs to be some mechanism to temporarily override that to listen to the takes.

[ferropop]

What this breaks is the current (nice) ability to quickly cycle takes directly into Melodyne.

This is where the folder heirarchy could come into play. You could (directly in the Takes folder) cycle takes and have it promote up to the MELODYNE folder, in realtime changing the active take - just like Reaper currently does. And the MELODYNE folder auto promote itself up to PARENT (AUDIBLE), while you are cycling.

[ferropop]

...and this is why well-thought-out, "handcuffed" comping workflows are standard for these extremely complex situations, because this all gets insane so quickly.

[bolgwrad]

Here's an idea, which is probably possible with custom actions already.

A new fixed-lane track, or track switched to fixed lane, by default inserts 'Lane zero', so two lanes are displayed from the off. (If converted back it becomes 'Lane one').

Any lanes subsequent to Lane Zero are children of the Lane Zero 'folder lane'.

Just flying it up the flagpole, see if anybody salutes.

[Subz]

Here is my 2p,

1) can I turn it off/not use it (stay how it is)

2) if I make a mistake and add a lane, can I easily remove it/turn it off (and go back to the old way)

Not saying I will never use it!

But I have a workflow (that this could improve! But would like to adapt in my own time)

Subz

[bolgwrad]

I have Ctrl+Space set to "Track properties: Fixed item lanes' and Win+Space set to "Free item positioning". They toggle, and don't cause any workflow problem.

[edit] Bear in mind that items assigned to lanes maintain their fixed-lane status, so if you were to assign a particular media item to lane 2 in a track, and drag it to another track it will force the target track to be fixed lane also. (Except it will drop into lane 1, which means switching that track back to non-lane mode won't affect the display).

[IrishRover79]

Originally Posted by MonkeyBars

Not sure where you get that idea at all. Have you tried them out in the dev versions? Fixed lanes are totally agnostic, like subtracks. Nothing at all says "different recordings of the same thing" in their design to me.

Yes, I have tried fixed item lanes out, including in the latest version (reaper673+dev1227_universal). It seems pretty clear to me from the design that the devs are thinking of fixed item lanes primarily as an alternative to the take system. They're trying to build a different way to record material to tracks and to comp that recorded material using razor editing, which will sit alongside the existing take system. That why there's an option to record to lanes rather than to takes, and why you can use razor editing to build a comp.

I'm not saying you couldn't use fixed item lanes as track versions, just that that doesn't seem to be what the devs have in mind currently. If it were, they'd make it possible to show only one lane at a time and give us keyboard shortcuts/actions to move backwards and forwards through the lanes. There would also be option to convert lanes to versions and versions to lanes.

In fact, lanes and versions COULD complement each other very nicely. That's how it works in Logic and you can probably do the same thing in Cubase as they're very similar in this respect. You record your takes. Then you build your comp. THEN, you can duplicate that 'track alternative' to a new one and 'flatten and merge' your comp to one item/region. Track alternative B is your clean audio with no splits. But track alternative A is your backup holding all your original takes and your comp, which remains editable in case you need to go back and adjust it.

[ferropop]

Sort of where I was going with the folders idea -- as a super-open-ended way to do versioning. Just have a dedicated place to hold them, and the ability to choose what is audible.

It's not the same (of course), I just think if versioning was on the devs' mind they would have been putting in the nuts and bolts early in the process - and so why not try to refine where things appear to be going with the info we've been given.

[Gass n Klang]

Mh, the folders thing is way too complicated. Think about a 40 channel orchestra recording. Gets messy pretty quickly if you've got all these tracks. Versioning seems to be a good alternative.

[IrishRover79]

At least BirdBird's track versions script seems to play nice with fixed item lanes...

[ferropop]

It doesn't work with Melodyne

Save for this, it is an incredible script.

[ferropop]

Here's I think the fundamental divide, that causes a ton of complication and head-butting...

It's the fact that Multiple Lanes CAN play at once.

This "freedom" really creates some serious hiccups in trying to create any sort of sensible workflow when comping specifically. Of course there's legitimate/creative/amazing uses for a "laned FIP", it's just wayyyy too unrestricted for the specific task of comping.

If there were a Comping Mode, similar to Lanes, similar to FreeItemPositioning mode, but with the restriction of "only one lane can be audible", it'd be much easier to enforce a workflow - and all the actions necessary for comping happen naturally.

ie, functionality to set a specific lane as the "audible lane".
-it could be a COMPED lane (items committed from swipe-comping takes). Store multiple comps, put the desired ones on an Audible Lane.
-or you could choose a Folder of Takes as audible (during comping), and Solo Markers would have to operate "On Lane Folders" instead of the entire track.




Also you could collapse it and the collapsed state is obviously The Audible Lane.

...you can even imagine putting all the COMPS in a folder, and then Solo Markering those - comping the comps.